JPH05224822A - Collective storage device - Google Patents

Abstract

PURPOSE:To set logical device constitution without considering physical size and to easily set the logical device constitution. CONSTITUTION:The collective storage device 108 which consists of plural physical storage devices 108 and is read and written from a host device is equipped with a physical device constitution information table 110 which contains constitution information on the physical storage devices 108, a logical device constitution information table ill which contains logical device constitution information when the physical storage devices 108 are divided into plural logical devices, and division specifying means 102 and 103 which perform at least one of (n)- equal-division specification for equally dividing the total storage capacity of the physical storage devices 108 into (n) logical devices, division distribution ratio specification for performing division specification by specifying the number of divisions and the distribution ratio of the respective logical devices, and physical border division specification for specifying division positions in physical storage device units according to a command from the host device.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention makes it appear that a plurality of storage devices such as a disk array device are connected to a host device (host), and a plurality of logical devices different in physical configuration are connected to the host device. The present invention relates to a method of defining a logical device in a collective storage device capable of performing.

[0002]

2. Description of the Related Art Japanese Laid-Open Patent Publication No. 2-236714 discloses an array type disk drive mechanism system. In this system, it is possible to flexibly control the logical configuration. Further, Japanese Patent Laid-Open No. 2-239348 discloses that
A method of dividing one physical storage medium into a plurality of logical devices is disclosed.

[0003]

DISCLOSURE OF THE INVENTION In the above-mentioned prior art, a disk array device or the like which is made up of a plurality of storage media and can be made to appear as if a plurality of logical devices having a configuration different from the physical configuration are connected to a host device. No consideration was given to simplification of the method of setting the logical device configuration in the collective storage device, and there was a problem that the designation of the logical device configuration becomes complicated.

Further, when the storage means for storing the logical device configuration information fails and the device configuration information is lost, the information stored in the collective storage device cannot be read correctly. ..

The present invention, like a disk array device,
In a storage device that is composed of multiple physical storage devices and has a function that can be used by dividing it into one or more logical devices, when designating the logical device configuration, specify only the logical information without considering the physical configuration. It is to provide a collective storage device capable of

Another object of the present invention is to provide a collective storage device which can prevent the recorded data from becoming unreadable due to loss of logical device configuration information due to hardware failure.

[0007]

A collective storage device according to the present invention is a collective storage device composed of a plurality of physical storage devices, and the number of partitions n ( (n is an integer of 2 or more) is provided to divide the total storage capacity of the physical storage device into n logical units by dividing the total storage capacity into n equal logical units.

Another collective storage device according to the present invention is, in a collective storage device composed of a plurality of physical storage devices, a division number and a distribution of each logical device as a division designation means to a plurality of logical devices. A division distribution ratio designating means for designating a ratio by designating a ratio is provided.

Still another collective storage device according to the present invention is a collective storage device composed of a plurality of physical storage devices, wherein the physical storage device unit is used as a unit for designating division into a plurality of logical devices. Further, a physical boundary division designating means for designating the division position is provided.

According to another aspect, the collective storage device according to the present invention comprises a plurality of physical storage devices, and in a collective storage device read and written by a higher-level device, A physical device configuration information table that stores configuration information; a logical device configuration information table that stores logical device configuration information when the plurality of physical storage devices are divided into a plurality of logical devices; According to the command, the total storage capacity of the physical storage device is n
Designation of n equal divisions into n equal logical units, designation of division distribution ratio by designating the number of divisions and distribution ratio of each logical device, and division position for each physical storage device. And a division designation means for designating at least one of physical boundary division designations for designating.

In this collective storage device, preferably, the physical device configuration information table and the logical device configuration information table are provided on a battery backup memory. The contents of the logical unit configuration information table are also stored in another non-volatile memory.

Contents of each of the physical device configuration information table and the logical device configuration information table when another physical storage device is added to the physical storage device for which the logical device configuration information is determined. A table updating means for automatically incorporating a new physical storage device into an existing logical device configuration may be further provided by modifying a part of the above.

[0013]

According to the present invention, when designating a logical device configuration for a collective storage device, instead of designating it based on physical level information, "n equal division designation" and "division distribution ratio designation" are performed. , "Physical boundary division designation", etc. are allowed at a high level. The information specified at this high level is automatically converted to the physical level in the collective storage device,
The logical device configuration is defined (that is, the contents of the logical device configuration table are set).

Further, in order to prevent the loss of the logical device configuration information, the logical device configuration information is duplicated and held in different kinds of storage means. As a result, when the information on one of the storage units cannot be read due to a hardware failure, the logical unit configuration information is lost and recorded by reading and using the information on the other storage unit. It is possible to prevent that data cannot be read.

[0015]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below by taking a disk array device as an example.

FIG. 2 shows the physical configuration of the disk array device according to this embodiment together with the host device. The disk array device 203 includes a plurality of small disk devices 108, an internal bus 206, a control device 202, and a host device interface 205, and is a SCSI (Small) device.
Computer System Interfac
It is connected to the higher-level device 201 by the higher-level device interface 205 such as e) and reads / writes information. Here, the read / write data 207 from the higher-level device 201 is divided into small disk read / write data 208 by the control device 202, and the small disk 108 is written and read. Therefore, the disk array device 203 has a feature that data can be read and written at high speed because writing and reading to and from different small disks 108 can be simultaneously performed in parallel.

FIG. 9 shows a more detailed relationship between physical blocks and data arrangement. As shown in the drawing, each physical drive is divided into a plurality of physical blocks, and serial block numbers are assigned to the physical blocks of all physical drives so as to sequentially circulate through each physical drive. That is,
The head data is stored in the first block 909 of the first drive (physical drive) 901, and the following data is stored in the second block 910 (the first block in the second drive 902) of the second drive 902, and the following. The data is stored in the third block in the third drive 903, and so on.

Therefore, the disk array device 203 logically has one large capacity disk 3 as shown in FIG.
It can be handled as 01 by the higher-level device 201.
It is also possible to divide this one logically large capacity disk into several logical disks 401, 402, ... As shown in FIG. As described above, in a collective storage device that is configured by a plurality of storage devices and can appear to be connected to a host device as a plurality of logical devices having a configuration different from the physical configuration, it is necessary to set the logical device configuration. ..

Next, a method of setting the logical device configuration will be described. Various methods can be used to set the logical device configuration, such as providing a function to set the logical device configuration offline. Here, however, a command for defining the logical device configuration is provided to the host device. The method of receiving the logical unit configuration information as the parameter of will be described as an example.

FIG. 11 shows an example of the format of commands and parameters transferred from the host device to the disk array device. As shown in (a) of the figure, the command includes a command code section 2 indicating the type of command for performing "logical device configuration definition" in addition to the normal "write" and "read", and the length of the data to be transferred. Data length part 3 that specifies
And a part 4 of other accompanying information. The parameter (data) of the split command is, as shown in FIG. 2B, composed of a designation method section 6 indicating which of a plurality of designation methods to be described later for division and its parameter 7.

FIG. 1 shows a detailed configuration of the control device 202 shown in FIG. In FIG. 1, the control circuit 101 includes a microprocessor 102, a program storage memory 103, and a data buffer 104, and controls the entire disk array device. In response to an instruction from the control circuit 101, the higher-level device interface circuit 105 exchanges commands and data with the higher-level device. Battery backup memory 109
Stores the physical configuration information table 110 and the logical device configuration information table 111 in this device. The small disk interface circuit 107 also exchanges data with the small disk 108 based on an instruction from the control circuit 101.

Next, the operation when the logical unit configuration definition command is issued from the host unit will be described in detail. The control circuit 101 activates the host device interface circuit 105 after the device starts up, and waits for a command to be issued from the host device. When a command is issued from the host device, the host device interface circuit 105 delivers the command to the reception control circuit 101. The control circuit 101 is
When the command is received and the command code is analyzed and it is recognized that the command is the logical unit configuration definition command, the command and the parameter information are delivered to the logical unit configuration definition processing program (described later in FIG. 5).

FIG. 6 shows the contents of the parameters passed from the host device in the logical device configuration definition command.
The parameter delivered from the higher-level device together with the logical unit configuration definition command has a field 601 indicating the “designating method” of the logical unit configuration definition, as shown in FIG. The contents of this field 601 are set in the designation method section 6 of FIG. This field indicates whether the method of designating the logical configuration definition in this command is "physical size designation", "n equal division designation", "division distribution ratio designation", or "physical boundary division designation". The fields after “designation method” are the parts to be set in the parameter 7 of FIG. 11B, and differ depending on the content of the designation method.
“Specify physical boundary size” 601a is the logical unit number 60
2 and its physical size 603 are designated. "N
The “equal division designation” 601b is a method of designating only the division number 604. According to this designation method, the total physical capacity of the disk array device can be evenly divided into the same number of logical devices as the number of divisions by simply specifying the number of divisions without specifying the physical size of each logical device. "Specify allocation ratio" 601
c is the number of divisions 605, the first logical unit allocation ratio 606, and the second
The logical device distribution ratio 607 and the third logical device distribution ratio 608 ...
Is a method of specifying. According to this designation method, if the number of divisions is three, the distribution ratio of the first logical device is "2", the distribution ratio of the second logical device is "1", and the distribution ratio of the third logical device is "3". In this way, the capacity ratio of each logical device to the whole can be designated. "Physical boundary division designation" 601
d is a method of designating the number of divisions 609, the first physical boundary position 610, the second physical boundary position 611, the third physical boundary position 612, .... According to this designation method, each physical boundary position to be divided into physical device units can be designated directly.

FIG. 7 shows a battery backup memory 109.
The format of the physical configuration information table 110 stored above is shown. The physical configuration information is information set when the device is assembled or a magnetic disk is added, and the number of connected magnetic disks 701 and the number of magnetic disk columns is 7
02, the number of magnetic disks per column 703, a single magnetic disk capacity 704, and the total physical capacity 705. The “column” indicates the arrangement of magnetic disk groups connected by the internal bus 206.

FIG. 8 shows the logical unit configuration information table 111.
Shows the format of. The logical device configuration information is information that is normally set by the logical device configuration definition command when the disk array device is installed. For each logical device, the logical device capacity 801 and the logical device start drive N are set.
o. 802, logical unit start logical block No. 80
3, logical unit end drive No. 804 and the logical block end logical block No. 804. Each field has 805 fields.

FIG. 5 shows a flow chart of the processing of the logical unit configuration definition command. In this process, the content of the designation method field 6 in the parameter shown in FIG. 11B is referred to (step 501), and if “physical boundary size designation”, then on the logical unit configuration information table 111 (FIG. 8). The capacity field 801 of the field corresponding to the logical unit number
The content of the physical size 603 (FIG. 6) is set in the field No. 802, starting logical block No. 803, end drive No. 804, end logical block No. 805 is calculated and set (step 502).

Next, if the designation method is "n-divided designation" (step 503), the physical device configuration information table 110
Of the total physical capacity field 705 of the logical unit configuration information table 11
1 is set in the capacity field 801 in the first drive field, and the start drive No. 802, starting logical block No. 8
03, end drive No. 804, end logical block No. 805 is calculated and set (step 504).
This operation is repeated the number of times indicated by the division number field 604.

If a surplus occurs when the content of the total physical capacity field 705 in the physical device configuration information is divided by the number of divisions 604, a value obtained by discarding the surplus is the capacity in the logical device configuration information. Set in the field 801. In the processing shown in step 504 of FIG. 5, the capacities of the divided logical devices are all of a uniform size, and the area of the surplus capacity is not allocated to any logical device and becomes an invalid area. Alternatively, the final partition logic unit may include the surplus.

Further, when the designation method is "division distribution ratio designation" (step 505), the following is performed. In the case of “designation of division distribution ratio”, for example, when allocating 10% of the total physical capacity as the first logical device and allocating the rest as the second logical device, “2” is designated as the number of divisions 605, and the first logical device allocation The ratio 606 is “1”, but the second logical unit distribution ratio 60
"7" is designated as "7", and "0" is designated after the third logical unit distribution ratio 608. Therefore, the content of the total physical capacity field 705 is taken out and multiplied by each distribution ratio to obtain the capacity value of the logical device, and the logical device configuration information table 1 is obtained.
It is set in the capacity field 801 in 11. Then, in the same manner as above, the start drive No. 802, starting logical block No. 803, end drive No. 804, end logical block No. 805 is calculated and set (step 506). This operation is repeated the number of times indicated by the division number field 605.

Finally, the designation method is "physical boundary division designation".
If not (step 507), a designation error of "designation method" is reported to the host device (step 509). If the designation method is "physical boundary division designation", the following processing is performed. Here, the "physical boundary division designation" means division in each arrangement of the magnetic disks in the column direction in the configuration shown in FIG. This division is advantageous for multiple operations because one magnetic disk is not divided into a plurality of logical devices. When this designation is made, for example, two columns as the first logical device,
When allocating one column as the second logical device, “2” is set in the division number field 609, “2” is set in the first physical boundary position field 610, and the second physical boundary position field 61 is set.
“1” is set in 1 and “0” is set in the third physical boundary position field 612 and subsequent fields. In this case, the contents of the first physical boundary position field 610, the number of magnetic disks per column field 704, and the single magnetic disk capacity field 70
The product of 4 is set in the capacity field 801 in the logical unit configuration information. Then, in the same manner as above, the start drive No.
802, starting logical block No. 803, end drive No. 804, end logical block No. 805
Is calculated and set (step 508). This operation is repeated the number of times indicated by the division number field 609.

In the above-mentioned "physical boundary size designation", "n equal division designation", and "division distribution ratio designation", an example of dividing in the direction crossing the physical drive is shown, but instead of this or in addition to this. In addition, it is possible to divide the physical drive unit. When the division is performed for each physical drive, the data transfer rate to the higher-level device is lower than that in the above case, but the operation can be performed without being affected by other physical drives, so that the multiple operation performance is improved.

Further, although four designation methods are shown in FIG. 6, the present invention need not have all of these designation methods.

The above is the outline of the operation when the logical unit configuration definition command is received from the host system. In this embodiment, in order to facilitate setting of the logical unit configuration information,
It has a function to automatically set the logical unit configuration information when adding a magnetic disk. The flow of this update processing is shown in FIG.

First, when the magnetic disk 108 is added, the capacity of the added magnetic disk is calculated and the contents of the physical device configuration information table 110 (FIG. 7) are updated (121). Then, in order to update the contents of the logical unit configuration information table 111, the capacity of the additional magnetic disk is added to the capacity field 801 of the last valid logical unit (122). Next, the end drive No. of the last valid logical device is set. 804 additional magnetic disk drive No. Is set (123). Finally, the end logical block No. of the last valid logical device. The end logical block No. of the additional magnetic disk is added to 805. Is set (124). Needless to say, the re-setting can be performed by the above-mentioned setting method without depending on this function at the time of expansion.

The logical unit configuration information set by the above method is stored in the battery-backed memory 109.
It is written in the upper logical device configuration information table 111.
This information is used to obtain the physical address on the magnetic disk from the logical address specified by the host device 201 when accessing the magnetic disk 108. Therefore,
If this information is lost, the magnetic disk 108 cannot be accessed. Generally, important information is duplicated and held, but considering hardware failure, duplication of the same hardware is required. In this embodiment, instead of duplicating the hardware, the logical device configuration information on the battery backup memory is also written on the floppy disk 112 via the FD interface circuit 106. Instead of the floppy disk 112, an area may be secured on the magnetic disk 108 and written there.

In the above embodiments, an example of a disk array device using a magnetic disk as a storage device is shown, but the same can be realized when another storage medium is used.

[0037]

According to the present invention, the logical device configuration can be set without being aware of the physical size, and the logical device configuration can be easily set.

By duplicating the logical device configuration information in different hardware, it is possible to prevent loss of the logical device configuration information due to a hardware failure and unreadable data being recorded.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a collective storage device according to an embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration example of a disk array device to which the present invention is applied.

FIG. 3 is an explanatory diagram of a logical configuration example of the collective storage device of FIG. 1.

FIG. 4 is an explanatory diagram of another logical configuration example of the collective storage device of FIG.

FIG. 5 is a flowchart of processing of a logical unit configuration definition command in the embodiment.

FIG. 6 is an explanatory diagram of a parameter format in a logical unit configuration definition command according to the embodiment.

FIG. 7 is an explanatory diagram of a physical device configuration information table according to the embodiment.

FIG. 8 is an explanatory diagram of a logical device configuration information table according to the embodiment.

FIG. 9 is an explanatory diagram of physical blocks and data arrangement according to the embodiment.

FIG. 10 is a flow chart of an automatic update process of table contents when a disk device is added in the embodiment.

FIG. 11 is an explanatory diagram of a command and data format from a higher-level device in the embodiment.

[Explanation of symbols]

101 ... Control circuit, 102 ... Microprocessor, 10
3 ... Program memory, 104 ... Data buffer, 1
05 ... Host device interface circuit, 106 ... Floppy disk interface circuit, 107 ... Small disk interface circuit, 108 ... Small disk, 109 ...
Battery backup memory, 110 ... Physical configuration information table, 111 ... Logical device configuration information table, 201 ...
Host device, 202 ... Control device, 203 ... Disk array device, 205 ... Host device interface, 206 ... Internal bus, 207 ... Read / write data from host device, 208
... small disk read / write data, 301 ... logical device, 4
01 ... Logical device.

Claims (8)

[Claims] 1. In an aggregate type storage device composed of a plurality of physical storage devices, the number of divisions n (n
Is an integer greater than or equal to 2) to divide the total storage capacity of the physical storage device into n equal logical divisions into n logical units. apparatus. 2. In a collective storage device composed of a plurality of physical storage devices, as a partition designation means for a plurality of logical devices, partition designation is performed by designating the number of partitions and a distribution ratio of each logical device. 2. The collective storage device according to claim 1, further comprising divisional allocation ratio designating means. 3. The collective storage device according to claim 1, wherein each physical storage device is divided into the number of divisions. 4. In a collective storage device composed of a plurality of physical storage devices, a physical boundary division designation for designating a division position for each physical storage device as division designation means for a plurality of logical devices. A collective storage device comprising means. 5. A physical storage device configuration information table configured to store configuration information of a plurality of physical storage devices in a collective storage device configured from a plurality of physical storage devices and read and written by a host device, Logical device configuration information table that stores logical device configuration information when the physical storage device is divided into a plurality of logical devices, and the total storage of the physical storage device according to a command from the host device. An n-partitioned designation that divides the capacity into n logical devices divided into n equal parts, a divisional allocation ratio designation that specifies the number of divisions and the allocation ratio of each logical device, and the physical storage device unit And a partition designation unit for designating at least one of physical boundary partition designations for designating a partition position. 6. The collective storage device according to claim 5, wherein the physical device configuration information table and the logical device configuration information table are provided on a battery backup memory. 7. The collective storage device according to claim 6, wherein the contents of the logical device configuration information table are also stored in another non-volatile memory. 8. Each of the physical device configuration information table and the logical device configuration information table when another physical storage device is added to the physical storage device for which the logical device configuration information has been determined. 6. The collective storage device according to claim 5, further comprising table updating means for automatically incorporating a new physical storage device into an existing logical device configuration by modifying a part of the contents of the above. ..